The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The README and installation scripts include commands that download and execute shell scripts directly from the author's GitHub repository. Specifically, curl -fsSL https://raw.githubusercontent.com/yousefabdallah171/code-quality-reviewer/main/install.sh | bash and a corresponding PowerShell command using iex are provided for automated setup.
[EXTERNAL_DOWNLOADS]: The skill's installation instructions involve fetching source code and wrappers from the author's public GitHub repository (yousefabdallah171/code-quality-reviewer) using git clone or npx commands.
[COMMAND_EXECUTION]: The workflow instructions (SKILL.md and subskills) direct the agent to execute a central Python orchestrator (scripts/codeguard_workflow.py) to run scans, generate reports, and create feature specifications. These scripts perform local file system operations within the user's project directory.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external project code, which constitutes a surface for indirect prompt injection as this untrusted data is processed by the scanning engine.
Ingestion points: Project source files (e.g., .js, .py, .go, .php) are read by the scanning scripts in the scripts/ directory.
Boundary markers: The skill does not explicitly use boundary markers or delimiters when presenting findings derived from user code to the agent.
Capability inventory: The skill can execute Python scripts, write files to the .codeguard/ directory, and influence the agent's planning and implementation tasks.
Sanitization: Content from project files is processed using regular expressions to identify patterns; the findings are then summarized in Markdown reports produced by the Python scripts.

codeguard